Offshoring in the Semiconductor Industry: a Historical Perspective Clair Brown and Greg Linden U.C

Prepared for the 2005 Brookings Trade Forum on Offshoring of White-Collar Work Draft (May 1, 2005) Offshoring in the Semiconductor Industry: A Historical Perspective Clair Brown and Greg Linden U.C. Berkeley May 2005 Abstract: Semiconductor design is a frequently-cited example of the new wave of offshoring and foreign-outsourcing of service sector jobs.1 It is certainly a concern to U.S. design engineers themselves.2 In addition to the current wave of white-collar outsourcing, the industry also has a rich experience with offshoring of manufacturing activity. Semiconductor companies were among the first to invest in offshore facilities to manufacture goods for imports back to the U.S. A brief review of these earlier manufacturing experiences and their impact on the fortunes of the domestic industry and its workers can help to illuminate the current debates over offshoring in services. Because meaningful data about the impact of the offshoring of chip design (and even manufacturing) are limited, we rely on a more qualitative analysis for our key points. We have conducted dozens of interviews with engineers and managers at numerous semiconductor and related companies in the United States, Asia, and Europe over the past six years. Our research also incorporates the rich store of publicly available information in trade journals and company reports. This paper describes the two previous stages of offshoring semiconductor assembly jobs and of outsourcing semiconductor manufacturing and the impact they had on the U.S. semiconductor industry. We argue that the initial concern about losing domestic jobs in both stages turned out to be unfounded as the industry used the situation to its competitive advantage by becoming cost competitive (assembly stage) and by developing the fabless sector (manufacturing stage). We then analyze the on-going stage of offshoring design jobs, and compare this stage to the two that came before in order to explore the possible impact on domestic jobs and the U.S. semiconductor industry. We begin in section one with a brief description of the stages of semiconductor production and our analytical framework. Section two looks at the offshoring of assembly jobs, and section three analyzes the foreign outsourcing of manufacturing. Section four explores the offshoring of design jobs, and concludes with a discussion of what this means for the U.S. Acknowledgments The authors would like to thank the Alfred P. Sloan Foundation, the Institute for Technology, Enterprise and Competitiveness (ITEC/COE), and Omron Fellowship for funding, and the Institute of Industrial Relations, UC Berkeley, for administrative support. They are also grateful to Ben Campbell, Michael Flynn, Ron Hira, Dave Hodges, Rob Leachman, Elena Obukhova, Devadas Pillai, and Bill Van Der Vort for their valuable contributions, and to Melissa Appleyard, Hank Chesbrough, Jason Dedrick, Rafiq 1 See for example “The New Global Job Shift,” Business Week, February 3, 2003, cover story and “Another Lure Of Outsourcing: Job Expertise,” Wall Street Journal, April 12, 2004; Page B1. 2 “2004 Salary Survey: It's an outsourced world, EEs acknowledge,” EE Times, August 27, 2004. Brown and Linden - 2 Dossani, Richard Freeman, Deepak Gupta, Bradford Jensen, Ken Kraemer, Frank Levy, and Tim Sturgeon for helpful discussions. The authors are responsible for any errors. I. Introduction: The Industry and Analytical Framework In order to understand the process of outsourcing of activities in the semiconductor industry, we begin by describing the stages of production. The most important type of semiconductor, and the one which concerns us here, is the integrated circuit, or “chip,” which is basically a network of tiny wires fabricated on a surface connecting transistors that switch on and off for processing data in binary code. The development and manufacturing of semiconductors involve three primary activities in the value chain: design, fabrication (front end), test and assembly (back end). During design, the desired electronic circuits progress through a series of abstract to physical representations. During fabrication, the circuits of the chips are built up on the surface of a flat, round silicon wafer in successive layers Assembly is, typically, the process of cutting the wafer into individual chips (or die), which can number up to 1000, depending on die size, and packaging the delicate chip in a protective shell that includes connections to other components. The semiconductor industry has successively undergone phases of offshoring—first assembly, then fabrication, and now design. The economic characteristics of each step of the process differ significantly. Design is skill intensive, and requires expensive EDA (electronic design automation) tools, which tend to be licensed per design engineer Fabrication required a huge fixed investment (currently on the order of $2 billion) to build a plant (called a fab) that holds a wide variety of expensive equipment and that meets extreme requirements of cleanliness. Assembly requires expensive test and assembly equipment, but the overall costs of plant and equipment are much lower than for the fab. Overall, worker skill requirements go up along the value chain. Semiconductors can be cost-effective to offshore in any location with adequate transportation facilities because their very high value-to-weight ratio reduces the penalty for long hauls between factory and customer or between stages of production. However, equipment costs dominate labor costs, especially for fabrication, and this has limited the attractiveness of low-cost labor locations. Even the most labor-intensive activity, chip assembly, has become more automated over time. As discussed below, other costs, including those relating to land, taxes, and government regulations, often affect decisions to offshore. The framework within which we will analyze the offshoring of the stages of the industry value chain relies on the concept of competitive advantage (Porter, 1985). A sustained advantage over rivals can be built on product (i.e., the intellectual property that defines functionality), price (i.e., the cost of production), or market attributes (i.e., new customers, customer service, brand reputation, and links to legacy products). These sources of competitive advantage provide the three principal reasons that firms offshore one of their activities: access to location-specific resources including engineering talent, cost reduction, and market development. When a firm with some non-imitable advantage moves an activity offshore to reduce its costs or improve access to resources, it improves (barring cases where the move is mismanaged) its competitive position against its rivals. In an expanding market like that for chips, the firm will grow and will hire more workers, some of whom will be in the home country and some offshore. However, some or all of the workers in the home country who were engaged in the activity that shifted offshore may lose their jobs, so that only the remaining home country workers benefit from the firm’s move offshore, along with the consumers of the lower-price products (see, for example, Garner, 2004). In addition, both exports and imports are increasing with market growth and the net impact of trade on jobs must be considered (Groshen, et al., 2005). Brown and Linden - 3 Numerous firm-level investments in a foreign location may change the location in such a way that it presents a new set of opportunities that lead to a transformation of the industry. A foreign location that is initially little more than a source of lower costs, especially labor, might develop over time as a specialized supply base. The changes can increase the value of the location to the point that the industry will eventually restructure around the new distribution of skills, and offshoring becomes the preferred mode for this activity. We will discuss below how this occurred for semiconductor assembly, but it has also taken place in other industries, such as hard disk drives (McKendrick, et al., 2000). II. Offshore Assembly: from offshoring to outsourcing Assembly was the easiest stage of production to be moved offshore. It was functionally separate from the other stages of production even when performed in close proximity to fabrication. Furthermore, assembly began with a relatively high use of less-skilled direct labor. During the 1980s, the US offshore companies switched to automation in response to a combination of increasingly intricate packaging requirements and higher equipment costs along with rising wages in some South-East Asian nations. As a result, the output per worker increased, and the typical plant still employs 1,000 or more workers. As of the mid-1990s, low-skilled workers made up about 80% of the staff of offshore assembly plants. The share of engineering and professional jobs was about 6%, and technicians made up another 13%.3 The move to offshore assembly led to a “hollowing out” of the US chip assembly sector, but kept the US chip industry cost-competitive as new rivals appeared in Europe and Japan. Over time, Asian suppliers appeared and took over a large portion of the business, so it went from offshoring to outsourcing, although most IDMs still own some assembly plants in Asia. The main lesson from this period of offshoring is that giving up one part of the value chain (at least as far as domestic production is concerned) may be necessary to “save” the domestic industry. The second lesson is that the initial moves offshore can have unforeseen dynamic consequences such as the emergence of foreign suppliers who dominate the industry segment. A. Offshoring, job loss, and competition Because of their high value-to-weight ratio, semiconductors could profitably be fabricated in the United States, air-freighted to Asia for assembly, and then returned to the United States for final testing and shipment to the customer. This system allowed the U.S. companies to take advantage of the specialized skilled and semi-skilled labor in the United States for design, fabrication, and key managerial functions while tapping the lower cost unskilled labor, land, and taxes of Asia for assembly.

Load more